IMS: Let Go of My WebRTC
Trying to tie WebRTC with IMS is a dreadful mistake.
IMS started as an architecture for mobile when 3G was still considered a future technology. It was positioned as the future architecture that would allow for easy third-party services integration in an all-IP world, along with many other goodies.
Fast forward some 10+ years from then: IMS is deployed by many service providers, but:
* Service providers mainly use the backbone side of IMS for their core network and not so much the services
* It was overcomplicated, so each vendor and service provider deployed it in a different flavor, making interoperability between them more of a challenge and requiring an SBC to bridge between networks
* The user didn't benefit from all those goodies promised, yet there are things such as QoS (Quality of Service) that are part of the IMS package that the user is not aware of.
In a nutshell, IMS didn't deliver on its promise. Service providers made significant investments in IMS, but the new and advanced services were launched in OTT islands and not over IMS. The main beneficiaries of IMS were the vendors selling IMS products. I worked in a company developing and selling technology software for IMS, so I witnessed this first-hand.
Fast forward to 2013. There is a new technology in the hype called WebRTC. WebRTC is a completely different story from IMS.
I recently attended the Upperside WebRTC conference in Paris, for which I also assisted in the planning. It was pretty clear that there are 2 camps: those that are still looking at the world through the prism of IMS, and those that already made the shift. The peak of this debate was presented in the panel, "WebRTC for the Operators: Opportunity or a New Threat?"
Basically some of the service providers and vendors focused on the gateway from WebRTC to IMS. In their view, IMS is the center of the communications world and all others (i.e. WebRTC) orbit around it.
In my view, this approach is the perfect path for telcos to miss yet another train.
In conclusion, trying to tie WebRTC with IMS is a dreadful mistake. I'm not saying there is no need for a gateway to connect between them--I even wrote a blog post about this, but as I said on the panel, just plug in a gateway and get done with it. There is no need to standardize this gateway in the 3GPP, and no need to make a big fuss about it. This is not the interesting thing. If I were a service provider, I would focus on building the platform for Web and application developers to add WebRTC-based services. This is much better off as an island than as an IMS thing.